A Prospective Study Evaluating Gender Differences of Serious Outcomes through Difficult Airway Physiological Score (DAPS) in the Emergency Department

Introduction Gender variation in critically ill adults after resuscitation is reported in many studies. However, this variation is not well established when evaluating the physiological instability in this population. This study aimed to prospectively evaluate the gender variation in serious outcomes by the difficult airway physiological score (DAPS) among critically ill patients requiring endotracheal intubation (ETI). Methods This is a cohort study conducted from August 2021 to December 2022 in the emergency department of Aga Khan University. The prospective validity of the difficult airway physiological score was derived using retrospective data and includes 12 variables: sex, age, time of intubation, hypotension, respiratory distress, vomiting, shock index >0.9, pH < 7.3, fever, anticipated decline, Glasgow Coma Scale (GCS) < 15, and agitation. The serious outcomes were cardiac arrest, mortality (within 1 hour after intubation in emergency), hypotension (systolic blood pressure <90 mmHg), and oxygen desaturation (SpO2 < 92%). The difference between males and females was assessed using the chi-square test, and the association of gender and serious outcomes was explored using Cox and logistic regression analysis. ROC curve analysis and area under the curve assessed score validity separately in males and females with serious outcomes. Results We enrolled 326 patients with a mean age of 50.3 (±17.8), with 123 (33.7%) females and 203 (62.2%) males. 198 (60.7%) patients were >45 years old, of which 136 (67%) were male and 62 (50.4%) female. Cardiac arrest was observed in 56 (17.2%), with 24 (19.5%) females and 32 (15.8%) males, p value 0.348. Hypotension after intubation was observed in 132 (40.5%) patients, 56 (45.5%) females and 76 (37.4%) males, p value 0.149. Oxygen saturation (<92%) was observed in 80 (24.5%) patients, 32 (26%) females and 48 (23.6%) males, p value 0.630. In females, the DAPS of 11 had an area under the curve of 0.863 (0.74–0.91). The sensitivity of the score was 84.8%, the specificity was 71.9%, the PPV was 77.8%, and the NPV was 80.4% with an accuracy of 78.9%. In males, the DAPS score of 14 had an area under the curve of 0.892 (0.57–0.75). The sensitivity of the score was 67%, the specificity 93.8%, the PPV 92.2%, and the NPV 72.2% with an accuracy of 79.8%. Conclusions The Difficult Airway Physiological Score (DAPS) predicts the risk of serious outcomes after intubation with high precision and reliability with different score cutoffs between the two sexes, highlighting the gender variation of a difficult airway.


Introduction
Airway management is a critical component of basic resuscitation and is crucial to patient safety and optimum clinical outcomes [1][2][3].Establishing and maintaining a patent airway to ensure adequate oxygenation and ventilation is the frst step of emergency medicine care [4].
However, the encounter with a difcult airway during induction or emergence poses signifcant challenges and requires a systematic approach for frst-pass success [5].Te timely identifcation of a challenging airway can allow emergency medicine physicians to anticipate potential diffculties and adopt appropriate strategies to mitigate risks and improve patient safety [6].Airway behavior in humans is infuenced by both biological (sex-related) and sociocultural (gender-related) determinants throughout their lifespan [7].Understanding these relationships is essential for interpreting gender-based anatomical variations and in studying their association with the occurrence of a difcult airway [7,8].
Traditionally, anatomical airway classifcations, such as the Mallampati score [9] or thyromental distance [10,11], have been employed to predict difcult airways.However, these scoring systems often fail to account for diferences that may occur specifcally in male or female patients.Recent studies have suggested that gender-related factors could play a crucial role in determining the likelihood of a difcult airway, prompting the need for a comprehensive investigation of these disparities [12,13].Te Difcult Airway Physiological Score (DAPS) is a useful tool designed to assess and predict the difculty of airway management before intubation.Te DAPS score was derived using retrospective data and includes 12 variables: sex, age, time of intubation, hypotension, respiratory distress, vomiting, shock index ≥0.9,pH < 7.3, fever, anticipated decline, Glasgow Coma Scale (GCS) < 15, and agitation.By assigning points to each parameter, DAPS allows clinicians to objectively assess the potential challenges they may face and to make informed decisions regarding the choice of airway management techniques and devices, ultimately ensuring better intervention and treatment of patients.
Tis study aims to explore and compare the determinants of difcult airway between male and female patients using the DAPS scoring system.By focusing on gender-based disparities, the study seeks to fll the current knowledge gap and provide information on how various factors of the DAPS score may predict difculties in airway management in a gender-specifc way.

Study Design and Setting.
A prospective cohort validation of DAPS was conducted in the emergency department (ED) of Aga Khan University Hospital from August 2021 to December 2022.Te recruiting center is an urban, academic, 62-bed emergency department that receives 60,000 patients annually.Te inclusion criteria of our study were all adult patients (≥18 years) who came to the ED and required endotracheal intubation (ETI).Patients with oropharyngeal tumors that require advanced airway measures due to distorted anatomy, patients with a history of cardiac arrest outside the hospital with ongoing CPR, and pregnant females due to varied physiological derangements were excluded from the study.Te intubation criteria were severe respiratory distress, worsening hypoxia that did not respond to noninvasive positive pressure ventilation, GCS less than 8, anticipated decline (intubation based on physician discretion), and imminent airway compromise.We estimated our sample size to be 268 based on an absolute precision of 6% with a 95% confdence interval and a 5% level of signifcance.Te sample size was calculated from a study by Smischney et al. [14] by the WHO calculator, showing a 52% rate of postintubation hypotension.Te initial calculation of the sample size relied on specifc assumptions, including expected efect size, variability, and anticipated dropout rates.However, during the study, a higher-than-expected enrollment rate of patients and lower dropout rates emerged, and the choice to enroll a greater number of patients than originally calculated was made to strengthen the robustness and applicability of our results.Tis decision was motivated by the increased statistical power derived from a larger sample, facilitating a more comprehensive exploration of gender diferences in serious outcomes using the DAPS in the emergency department.
Te Ethics Review Committee of Aga Khan University approved the study (ERC Number 2020-4975-14778).Consent was taken from the patient if he or she has intact capacity or from the accompanying attendant, who is the patient's decision-maker in the emergency department visit.

Data Collection.
Te triage nurses and the resuscitation room doctors identifed patients who needed intubation in the emergency department and informed the researchers.Te associates screened the patients after obtaining verbal consent from the patient (with intact capacity, which was understanding, appreciation, reasoning, and expression of choice about the process followed in the study) or the accompanying decision-maker, which was later followed by a written consent from either.Preintubation vitals at the triage were recorded followed by other demographic variables.During the collection of variables, the research associates did not interfere with the treatment of patients requiring ETI.Data were collected on a pretested questionnaire that was tested in data collection to derive the physiological score of difcult airways.Te data collected on the form were reviewed by the physician involved in the ETI to review missing data and confrm the data.Symptoms and vital signs at presentation, reason for intubation, difcult airway evaluation, drugs used in ETI, and other procedural data were collected.Te information collected was periodically reviewed by the principal investigator for accuracy.Te patient was followed in the emergency department 15 minutes and 1 hour after intubation for record of vitals, and the fnal disposition was recorded on an electronic medical record.Te estimated risk of serious outcomes associated with each level of score in the derivation study was not in the data collection form to prevent physicians from making treatment decisions based on the risk score.

Serious
Outcomes.Te primary outcomes were worsening hypotension and hypoxia.Hypotension was defned as a decrease in the systolic blood pressure (<90 mmHg), and hypoxia was defned as peripheral oxygen desaturation (<92%) within one hour of intubation.Secondary outcomes were cardiac arrest (defned as the absence of pulse after ETI in a critically ill patient in the ED) and mortality (defned as death occurring within 1 hour after intubation).All of the above outcomes were measured at diferent times: immediately after intubation, and 15 minutes and 1 hour after intubation.

2
Critical Care Research and Practice

Statistical Analysis.
Te study utilized Redcap for data entry and SPSS-22, along with Python 3.8.14, for analysis.Descriptive statistics were used for continuous and categorical variables, with comparisons made using appropriate tests (chi-square, Fisher's exact, t-test, or Mann-Whitney U test).Te association between gender and serious outcomes was explored using binary logistic regression models.Variables known or suspected to be associated with serious outcomes were examined with univariate binary logistic regression.Independent variables with a p value of <0.05 univariate regression were included in the multivariate model.Multivariate models were constructed using stepwise backward selection.Only variables with a p value of <0.05 were retained in the fnal model.Te goodness of ft was measured using the Hosmer-Lemeshow test, which evaluates the agreement between observed and predicted outcomes.Te Nagelkerke R square and overall correct classifcation percentage were also reported for each model.Te difcult airway physiological score was assessed using the ROC curve analysis, determining the AUC with 95% confdence intervals.Youden's J statistic identifed the main discriminating point of the DAP score, and sensitivity, specifcity, PPV, and NPV were calculated at various cutof points, all with 95% confdence intervals.A signifcant level of 0.05 was applied.

Results
In this study, 326 patients were enrolled, of which 123 were women and 203 were men.Te average age of the patients was 50.3 years, and the women had a slightly lower average age of 45.6 years compared to the men with an average age of 53.1 years.Te most common reasons for intubation were shortness of breath in 239 patients (73.3%) and then coma in 220 patients (67.5%), followed by respiratory distress in 165 patients (50.6%), anticipated decline in 143 patients (43.9%), hypoxia in 130 patients (39.9%), and metabolic acidosis in 89 patients (27.3%), followed by trauma with no signifcant diferences between the sexes.Most of the patients had a shock index below 0.9 (54.3% of the patients).
Table 1 presents the baseline characteristics of critically ill patients who required ETI.In general, the baseline characteristics of the critically ill patient, such as preintubation vitals, pH levels before intubation, and the HEAVEN criteria did not show signifcant diferences between sexes, except for age.
Table 2 shows the gender variation in the serious outcomes of the patients who required ETI with cardiac arrest, postintubation hypotension (SBP < 90 mmHg), and low oxygen saturation (<92%) with no signifcant statistical diferences between sexes.
Table 3 shows the univariate and multivariate logistic regression analysis to predict serious outcomes after ETI in men with shift duty (from morning to night from 8 am to 10 pm), shortness of breath, fever, drowsiness, trauma, others (unspecifed), hypoxia, anticipated decline, respiratory distress with signifcant association of serious outcomes in both univariate and multivariate logistic regression analysis, with isolated trauma, pH group <7.3, shock index >0.9,hypoxemia, extreme size, cardiac arrest, hypotension (SBP < 90 mmHg), and oxygen saturation (<92%) showing association only in univariate logistic regression analysis.
Table 4 shows univariate and multivariate logistic regression analysis to predict serious outcomes after ETI in women <45 years of age, with shift duty (from morning to night) from 8 AM to 10 PM, shortness of breath, fever, drowsiness, seizures, trauma, coma, altered mental status (GCS < 15), metabolic acidosis, respiratory distress, pH group < 7.3, and shock index > 0.9 were signifcantly associated with serious outcomes after ETI.
Te area under the curve (AUC) of 0.892 in Figure 1 and the AUC of 0.863 in Figure 2 suggest that the predictive model of the DAPS score used in this study has high precision in distinguishing male and female patients.Te sensitivity of the DAPS score of 11 in women was 84.8%, specifcity 71.9%, PPV 77.8%, and NPV 80.4% with an accuracy of 78.9%, while the sensitivity of the DAPS score of 14 in men was 67%, specifcity 93.8%, PPV 92.2%, and NPV 72.2% with an accuracy of 79.8%.

Discussion
Te Difcult Airway Physiological Score (DAPS) exhibits excellent accuracy, with a high AUC, in predicting which patients are likely to experience severe consequences after ETI, with a specifc score diferentiation between male and female patients.Te score has been developed and validated for difcult intubation in the emergency department, as a simple model that can be easily applied in clinical practice.A DAPS score of 11 in women has an accuracy of 78.9%, a sensitivity of 84.8%, a specifcity of 71.9%, a PPV of 77.8%, and an NPV of 80.4%, while a DAPS score of 14 in men has an accuracy of 79.8%, a sensitivity of 67%, a specifcity of 93.8%, a PPV of 92.2%, and an NPV of 72.2%.Te study has shown that the DAPS score can predict difcult intubation in the emergency and also reveals a high rate of severe morbidity related to difcult intubation.
Various scores have been suggested to assess the possibility of difcult airways in preoperative and intensive care unit (ICU) settings, like the LEMON score [15], which is the most commonly used tool to assess difcult airways, designed for use in the preoperative clinic setting before elective surgery.Although some of the LEMON criteria (such as the 3-3-2 rule and the Mallampati score) require an awake and cooperative patient, they lack guidance on the anticipated complications that an emergency airway may present.
Our score contrasts with other scores specifcally designed for ICU like the MACOCHA score [16], which predicts the difculty of tracheal intubation in ICU patients.Te score is calculated by assigning points to Mallampati class, apnea, cervical spine limitation, mouth opening, coma, hypoxemia, and nonanesthesiologist operator.A higher MACOCHA score indicates a higher risk of difcult intubation.In the primary study [17], the cutof of three or above rules out difcult intubation, provided a good negative predictive value of 97% and 98% and specifcity of 90% and 89%, with sensitivity of 76% and 73% but low positive predictive values of 48% and 36% in the original and Critical Care Research and Practice validation cohorts, respectively.In addition, there is no gender-based anatomical and physiological scoring, which, as our data suggest, has a signifcant impact on the parameters and outcome of intubations.
Furthermore, our data compared with the HYpotension Prediction Score (HYPS) [18] that predicts hypotension before and after intubation in the ICU setting, determining a total of 11 adverse hypotension factors after intubation, namely, increased age, APACHE II score, sepsis, intubation performed in settings of cardiac arrest or MAP 65 mmHg or decreasing SBP from 130 mmHg, acute respiratory failure, diuretics 24 hours before ETI, catecholamines or phenylephrine 60 minutes before ETI, and etomidate as sedative.Te score has a PPV of 11.9% and an NPV of 88.1% in the lowest category, and a PPV of 71.9% and an NPV of 28.1% at the highest risk threshold [14].However, this only serves as    On the other hand, the HEAVEN criteria [19] (hypoxemia, extremes of size, anatomical challenges, vomit/blood/fuid, exsanguination/anemia, and neck mobility problems) was the  [20][21][22] have been done to test the clinical signifcance of the criteria that found that the physiological factors of hypoxemia and exsanguinations are not associated with failure of frstattempt intubation.Te Difcult Airway Physiological Score (DAPS) expands on the parameters of diferent identifers of the difcult airway, resulting in a more comprehensive and superior predictability score.Te prediction of airway compromise is more imperative in an emergency setting than in a nonemergency setting.Te DAPS score provides a rapid and reliable prediction of difcult airways in the uncertain, severe, and urgent environment of the ED.It is also a novel score that takes into account several gender-based physiological diferences and provides distinct score cutofs for each gender in the interpretation, reinforcing the disparities between the airway behavior of the two genders.

Limitations
Although our study provides valuable information on gender diferences in serious outcomes related to difcult airway management through the DAPS score, it is important to acknowledge its limitations.As the study has been conducted in a specifc geographical area and in a single ED, it limits the generalization of the fndings to other settings.Te results obtained from a single institution may not be representative of the larger population.Additionally, there may have been a possible selection bias in the study sample.Triage systems are often used in emergency rooms to categorize patients according to the severity of their symptoms.Tis may result in overrepresentation of certain population characteristics and limit the use of fndings in contexts of other settings.Furthermore, the accuracy and reliability of the study scoring system may have been subjected to measurement bias as physiological parameters such as anticipated decline, agitation, and respiratory distress may be interpreted and recorded diferently between diferent medical professionals, resulting in inter-rater variability.Furthermore, in the study, not all possible confounders that could afect the results were taken into account.Te analysis could not adequately account for factors such as comorbidities, drugs, prior airway treatments, and the experience of the healthcare provider, all of which may have an impact on outcomes.Te ability to conclude the precise impact of gender on adverse outcomes could also have been limited by the lack of a comparison group, such as a control group without challenging airways.
Interpreting the results of our study requires an appreciation of these limitations, and future research should strive to address these issues to provide a more complete understanding of gender diferences in difcult airway management outcomes.

Table 1 :
Baseline characteristics of critically ill patients requiring ETI.
* Worsening type 2 failure, aspiration pneumonia, epileptic status, asthmatic status, severe agitation/diarrhea, impending threat to the airway due to burns or esophageal rupture or expanding hematoma.

Table 2 :
Variation in sex in the serious outcome of patients requiring ETI.

Table 3 :
Univariate and multivariate logistic regression analysis for predicting serious outcomes among men after ETI.

Table 4 :
Univariate and multivariate logistic regression analysis to predict serious outcomes among women after ETI.